RECONSTRUCTION OF RAT CALVARIAL DEFECTS WITH HUMAN ... · C Zong European Cells and Materials Vol....

109 www.ecmjournal.org

C Zong et al. Calvarial reconstruction with hMSC-PLGA constructsEuropean Cells and Materials Vol. 20 2010 (pages 109-120) DOI: 10.22203/eCM.v020a10 ISSN 1473-2262

Abstract

Human mesenchymal stem cells (hMSCs) can be used forxenogenic transplantation due to their low immunogenicity,high proliferation rate, and multi-differentiation potentials.Therefore, hMSCs are an ideal seeding source for tissueengineering. The present study evaluates the reconstructioneffects of hMSCs and osteoblast-like cells differentiatedfrom hMSCs in poly-lactic-co-glycolic acid (PLGA)scaffolds on the calvarial defect of rats. Two bilateral full-thickness defects (5mm in diameter) were created in thecalvarium of nonimmunosuppressed Sprague-Dawley rats.The defects were filled by PLGA scaffolds with hMSCs(hMSC Construct) or with osteoblast-like cellsdifferentiated from hMSCs (Osteoblast Construct). Thedefects without any graft (Blank Defect) or filled withPLGA scaffold without any cells (Blank Scaffold) wereused as controls. Evaluation was performed usingmacroscopic view, histology and immunohistochemicalanalysis respectively at 10 and 20 weeks aftertransplantation. In addition, fluorescent carbocyanine CM-Dil was used to track the implanted cells in vivo duringtransplantation. The results showed that while both hMSCConstruct and Osteoblast Construct led to an effectivereconstruction of critical-size calvarial defects, the bonereconstruction potential of hMSC Construct was superiorto that of Osteoblast Construct in non-autogenousapplications. Our findings verify the feasibility of the useof xenogenic MSCs for tissue engineering and demonstratethat undifferentiated hMSCs are more suitable for bonereconstruction in xenotransplantation models.

Keywords: Human mesenchymal stem cells, tissueengineering, xenotransplantation, bone reconstruction.

* Address for correspondence:Jinfu WangLaboratory of Stem Cells, Institute of Cell Biology,College of Life Sciences, Zhejiang University388, Yuhangtang Road, Hangzhou, Zhejiang 310058P. R. China

E-mail address: [email protected]

Introduction

There is a significant need for therapies to enhance healingin large bone defects caused by trauma, congenitaldeformity and tumor resection because each year over 1million cases of patients need bone graft procedures tocorrect such defects in the USA (Langer and Vacanti, 1993;Holtorf et al., 2006). Three primary approaches for bonedefect repair are autograft bone, allograft bone and tissueengineered bone transplantations (Kubler et al., 1995).However, transplantation of autograft bone or allograftbone is not optimal due to the limitation of supply and therisk of morbidity (Younger and Chapman, 1989; Hao etal., 2010). Bone tissue engineering, as a practical andpromising method of bone defect regeneration, hasbecome of major interest for bone reconstruction to repairlarge bone defects.

Mesenchymal stem cells (MSCs) derived from bonemarrow are an obvious source of autologous stem cells,and used as seeding cells for cell therapy and tissueengineering (Bruder et al., 1998a; Bruder et al., 1998b;Kon et al., 2000; Cui et al., 2007). MSCs have a strongregeneration potential and immunosuppressive propertiesthat are important for allografts. In fact, MSCs havealready been employed for clinical trials in a number ofcontexts, such as the facilitation of hematopoietic andimmune reconstitution after hematopoietic stem celltransplantation (Koc et al., 2000; Lazarus et al., 2005),construction of the vessel wall (Gong and Niklason, 2008),as well as the regeneration of cartilage using tissueengineering techniques (Jorgensen et al., 2004; Bernardoet al., 2007). Furthermore, a number of studies havedemonstrated that autologous MSCs cultured in scaffoldscan induce new bone formation in vivo and lead toimproved healing of critical-size defects (Yamada et al.,2004; Meinel et al., 2005; Mankani et al., 2006; Miura etal., 2006). In these studies, MSCs are always induced intoosteoblast-like cells with the osteogenic medium beforetransplantation. Osteoblast-like cells derived from MSCsshare similar characters with osteoblasts (Kassem et al.,1991). Osteoblast-like cells have a good ability of growthand proliferation and keep their biological characteristicsconstant after several passages. In addition, osteoblast-like cells can secrete much bone matrix to form maturebone with deposition of calcium salts, and play animportant role in osteogenesis and bone regeneration.

RECONSTRUCTION OF RAT CALVARIAL DEFECTS WITH HUMANMESENCHYMAL STEM CELLS AND OSTEOBLAST-LIKE CELLS IN

POLY-LACTIC-CO-GLYCOLIC ACID SCAFFOLDS

Chen Zong1, Deting Xue2, Wenji Yuan1, Wei Wang3, Dan Shen4, Xiangmin Tong4, Dongyan Shi1, Liyue Liu1, QiangZheng2, Changyou Gao3, and Jinfu Wang1*

1 Laboratory of Stem Cells, Institute of Cell Biology, College of Life Sciences, Zhejiang University, Hangzhou,Zhejiang 310058, P. R. China

2 Institute of Orthopedics, The Second Hospital, Zhejiang University, Hangzhou, Zhejiang 310009, P. R. China.3 Institute of Medical Materials, College of Material and Chemistry, Zhejiang University, Hangzhou, Zhejiang

310028, P. R. China4 Laboratory of Bone Marrow, The First Hospital, Zhejiang University, Hangzhou, Zhejiang 310006, P. R. China


C Zong et al. Calvarial reconstruction with hMSC-PLGA constructs

However, osteoblast-like cells lose some characteristicsand multipotential of differentiation as MSCs. Recentstudies also verify the ability of undifferentiated MSCs inscaffolds to enhance the bone formation and repair defectsof bone tissues in vivo (Korda et al., 2008; Niemeyer etal., 2010a). Therefore, it is interesting to compare the abilityof bone reconstruction between these two cell types.

In this study, we used porous poly-lactic-co-glycolicacid (PLGA) scaffolds as carriers. PLGA is a biodegradableporous material used commonly in tissue engineering. Itcan be removed from the body by normal metabolicpathways and is relatively harmless to the cells and tissues(Gopferich, 1996). We seeded hMSCs into scaffolds, andthen expanded and induced osteogenically hMSCs inscaffolds. Finally, the bone reconstruction ability ofundifferentiated hMSCs and osteoblast-like cellsdifferentiated from hMSCs in scaffolds was comparativelyanalyzed through the reconstruction experiment ofcalvarial defects in nonimmunosuppressed Sprague-Dawley (SD) rats.

Materials and Methods

Preparation of experimental materialsPorous poly-lactic-co-glycolic acid (PLGA) was fabricatedinto porous scaffolds by a porogen-leaching technique withgelatin particles as the porogen (Zhou et al., 2005). Briefly,the notable feature of this process was that the gelatinparticles bound together to form a three-dimensionalassembly through a water vapor treatment before thepolymer solution was cast. The PLGA scaffolds wereprepared as cylinders (10×3 mm) and have a porosity of85%. The pore diameter was about 280-450 μm.

Human mesenchymal stem cells (hMSCs) were isolatedfrom the whole bone marrow of three healthy donors atthe First Affiliate Hospital, Zhejiang University, using themethod described previously (Xie et al., 2005). All donorshad given informed consent before collection of their bonemarrow. hMSCs were cultured with minimal essentialmedium α (α-MEM; HyClone, Shanghai, China)supplemented with 10% fetal bovine serum (FBS; GibcoBRL, Hangzhou, China), 10,000 U/mL penicillin and10,000 U/mL streptomycin (Life Technologies, Beijing,China). hMSCs were incubated at 37°C in a high humidityenvironment containing 5% CO2. The medium wasreplaced every 3 days, and cells were passaged atapproximately 80% confluence. hMSCs in the thirdpassage were prepared for the following experiments.

Human calvarial osteoblasts (HCOs; Yuhengfeng,Beijing, China) were cultured according to themanufacturer’s instructions. Briefly, cells were seeded inT-75 flasks at a density of 5,000 cells/cm2 and culturedwith Osteoblast Medium (ObM, Yuhengfeng) composedof 500 mL basal medium, 25 mL FBS, 5mL osteoblastgrowth supplement (ObGS) and 5 mL penicillin/streptomycin solution (P/S) at 37°C and in a high humidityenvironment containing 5% CO2. After 24 hours of culture,unattached cells were discarded by refreshing the medium.Attached HCOs were continued to incubate under above

condition. The medium was replaced every 2 days, andcells were passed at about 80% confluence.

Two-month-old male non-immunosuppressed SD ratswith a body weight of 200 to 250 g were used as transplantrecipients for this study. The experimental protocol wasapproved by the Animal Research Centre at ZhejiangUniversity. Before surgery, the animals were kept inventilated, clean and standard air conditions at a constanttemperature of 22°C with a 12-hour light/day cycle.Animals could receive freely drinking water and standardlaboratory foods. All animals were cared for according tothe policies and principles established by State Scientificand Technological Commission (SSTC).

Cell seeding and culture of Constructs in vitroThe construct of PLGA-hMSCs (hMSC Construct) wasprepared as described previously with a small modification(Yang et al., 2010). Briefly, PLGA scaffolds were sterilizedusing 70% ethanol, followed by rinsing the scaffold severaltimes with phosphate buffered saline (PBS). hMSCs inthe third passage were suspended in α-MEM mediumwithout FBS at a cell density of 2×106 cells/mL. 250μL ofhMSC suspension was seeded into the scaffold by a syringelinked with the outlet of a screw-bottom, and then cellswere allowed to attach for 2h before 3mL of α-MEMmedium with 10% FBS per well was added into a six-wellplate for static culture. After 24h of static culture, theconstructs were transferred into a perfusion culture systemdescribed previously (Yang et al., 2010). The perfusionculture system was incubated at 37°C and in a highhumidity environment containing 5% CO2. The flow rateof medium was kept at 0.2 mL/min. The medium used forperfusion culture was the α-MEM medium supplementedwith 10% FBS (growth medium). The medium waspumped from the medium reservoir into the cell-scaffoldconstruct in each culture cassette, and then returned to themedium reservoir through the platinum-cured silicontubing. Growth medium in the medium reservoir wasreplaced at a rate of 50% every 2-3 days. The constructswere measured for DNA content after culture for 0, 6, 9,12 and 15 days, respectively. The optimal culture time wasobtained according to the DNA content analysis. Then,this time point was chosen as the preparation time in vitroof hMSC Construct before transplantation.

The construct of PLGA-osteoblast-like cells derivedfrom hMSCs (Osteoblast Construct) was prepared asfollows: Firstly, PLGA scaffolds seeded with hMSCs werecultured for 9 days in the perfusion culture system as inthe preparation of hMSC Construct. Then, the growthmedium was replaced with the osteogenic medium (α-MEM supplemented with 10% FBS, 50μg/ml ascorbicacid, 10mM sodium, β-glycerophosphate, and 10-8Mdexamethasone; Sigma, Shanghai, China) for another 14days of culture. Osteogenic medium was replaced every 4days. Then, before transplantation, the DNA content ofthe constructs was measured.

DNA content assayThe constructs cultured for 0, 6, 9, 12 and 15 days as wellas induced osteogenically for 14 days, were used for DNA



content assay. Three constructs per time point wereremoved from the perfusion system and stored in ddH2Oat -20°C until the assay was performed. For DNA contentassay, the frozen scaffolds with cells were thawed at roomtemperature and homogenized in 1mL lysis buffer (50mMTris/HCl, pH7.6, 0.1% (v/v) Triton X-100). The lysate wasassayed for DNA content using a fluorescent dye Hoechst33258. Briefly, the lysate was sonicated on ice for 30s andvortexed for 5-10s. After centrifugation, the supernatantwas collected. Standards of calf thymus DNA wereprepared in the range from 0 to 30μg/mL. TNE buffer(10mM Tris base, 1mM EDTA and 200mM Nacl) wasadded to each well of a 96-well plate at 50μL/well. Then,50μL standards and samples were added to each well intriplicate. Then, 100μL Hoechst 33258 dye solution (1μg/mL) was added to each well. Finally, the plate wasincubated for 10 min in the dark at room temperature, andthen read on Auto Microplate Reader (Infinite M200,Tecan, Austria; Ex350nm/Em450nm). Cell number wasdetermined by correlating DNA with a known amount ofcells. Samples were run in triplicate and compared againstcalf thymus DNA standards.

Assay of osteogenic differentiation in scaffoldsAlkaline phosphatase (ALP) activity, as an importantosteogenic differentiation marker, was detected to verifywhether hMSCs in PLGA scaffolds had differentiated toosteoblast-like cells. ALP activity assay was measuredusing an ALP measurement kit (Jiancheng BiotechnologyInstitute, Nanjing, China) with disodium phenyl phosphateas a substrate. Briefly, scaffolds with hMSCs undergoing9-day-culture in the growth medium were continued toculture in perfusion system with the osteogenic mediumfor another 14 days. Then, constructs (n=3) were movedfrom the perfusion system and washed three times in PBS,and homogenized in lysis solution (0.1% Triton X-100 inPBS) and sonicated for 5min. After the lysate wascentrifuged, 300μl supernatant was added into a 5mlCorning tube, and then 500μl buffer and 500μl substratesolution were added before incubation at 37°C for 15min.After that, the above mixture was added to1.5ml ofchromogenic reagent. Finally, the absorbance wasmeasured at 520nm, and the ALP activity of each samplewas calculated using the phenol standard curve.

RT-PCR was performed to determine the expressionof specific genes for osteogenesis in scaffolds. As above,the constructs (n=3) were collected and three times washedwith PBS. The constructs were pulverized and the totalRNA was extracted using Trizol reagent (TaKaRa,Shanghai, China) according to the manufacturer’sinstructions. RNA (1μg) was used to synthesize first strandcDNA by reverse transcriptase (Fermentas, Shanghai,China). The polymerase chain reactions (PCR) wereperformed to analyze the genes ALP, collagen I (COL I)and osteopontin (OPN) expressed by human osteoblasts.GAPDH was used as a housekeeper gene to normalize allmeasurements for the other products. Primers weredesigned using PRIMER EXPRESS 1.0a Software(Applied Biosystems) as follows: ALP: fwd 5’ ACA AGCACT CCC ACT TCA TC 3’, rev 5’ ATT CTG CCT CCT

TCC ACC 3’; COL I: fwd 5’ TCA AAG GCA ATG CTCAAA CA 3’, rev 5’ ACA TCA AGA CAA GAA CGA GGTAG 3’; OPN: fwd 5’ AGC CGT GGG AAG GAC AGTTAT G3’ , rev 5’ GGA GTT TCC ATG AAG CCA CAAAC 3’; GAPDH: fwd 5’ GAA GGT CGG AGT CAA CGG3’, rev 5’ GGA AGA TGG TGA TGG GAT T 3’. PCRreaction conditions were 94°C for 5 min, then 30 cyclesof 94°C for 30s, 53°C (ALP), 43°C (COL I), 52 °C (OPN)or 53°C (GAPDH) for 30s and then 72°C for 1 min, finallyfollowed by 72°C for 5 min. PCR products wereelectrophoresed on a 1.5% agarose gel and then stainedwith ethidium bromide. Each PCR product was run threetimes. Stained bands were visualized under UV light andphotographed.

Surgery and transplantation procedureAll surgery was performed under a protocol approved bySSTC. Round calvarial defects (5 mm in diameters, full-thickness) were created in two-month-old SD rats (Centerof Experimental Animals, Zhejiang University). Aftergeneral anesthesia with 4% chloral hydrate, the skin andunderlying tissues of vertex, including the periosteum andthe temporalis muscles, were raised to expose the calvaria,and then two full-thickness defects in symmetry to thesagittal suture were generated by a dental bur. Constantsaline irrigation was provided and the dura mater was keptintact with minimal invasion during the procedure. Theprocedure was performed under sterile conditions. Afterthe defects were implanted with grafts, a single layer ofsoft tissue was closed with absorbable sutures, and thenthe skin incision was closed with nylon sutures. Aftersurgery, SD rats were kept in the clean conditions withenough water and foods.

24 SD rats used for the experiment were divided intofour groups (see Table 1). The rats were sacrificed bycervical dislocation or euthanasia with an overdose ofchloral hydrate, respectively, at 10 and 20 weeks aftertransplantation, and their calvarias with grafts wereharvested.

puorG rebmuntaR tfeltatfarGtcefed

tatfarGtcefedthgir

A 6 CSMhtcurtsnoC

knalBtcefeD

B 6 CSMhtcurtsnoC

knalBdloffacS

C 6 tsalboetsOtcurtsnoC

knalBdloffacS

D 6 tsalboetsOtcurtsnoC

CSMhtcurtsnoC

Table 1. Treatment groups of experimental animals

hMSC Construct: the scaffold with hMSCs; OsteoblastConstruct: the scaffold with osteoblast-like cellsdifferentiated from hMSCs; Blank Defect: the defectwithout any graft; Blank Scaffold: the scaffold withoutany cells.



Cell tracing in vivoTo in vivo trace hMSCs and osteoblast-like cells inscaffolds grafted into rats, we used the fluorescentcarbocyanine CM-Dil to label cell membranes. The methodwas an improvement on the manufacturer’s protocol(Ferrari et al., 2001). Briefly, at 24h before transplantation,hMSCs or osteoblast-like cells in scaffold were labeledwith 4μg/ml CM-Dil for about 2h at 37°C, and then thescaffold was washed with PBS for three times andcontinuously cultured in the perfusion culture system untiltransplantation. Cell tracing was performed at 10 weeksafter transplantation. Tissue samples from defect areas wereharvested from 4 groups (2 samples per group) anddecalcified in Decalcifying Fluid (Zhongshan, Beijing,China) for 4 days at room temperature, and then washedin running water for about 2 hours. The 7μm frozen sectionof sample was prepared and then counterstained by DAPIfor 5 minutes. The result was observed using a ConfocalLaser Scanning Microscope (LSM-510, Carl Zeiss,Germany).

Histological evaluationTissue samples harvested from 4 groups (2-4 samples pergroup) at 10 and 20 weeks after transplantation,respectively, were fixed with 10% buffered formalin for24h and then several times washed with PBS. The samplewas the decalcified in neutral 10% EDTA solution for halfa month at room temperature, and then dehydrated in agraded alcohol series and embedded in paraffin. 5μmsections of the sample were prepared and stained withhematoxylin and eosin (H&E) and Modified MassonTrichrome as previously described (Garvey, 1984; Cowanet al., 2004). The stained sections were photographeddigitally under a microscope (Digital Camera DXM200F,Nikon, Japan). Semi-quantitative image analysis (Ahmadet al., 2004; Zsarnovszky et al., 2005;) was used to estimatethe new forming bone at 20 weeks after transplantation,i.e., to calculate the percentage of newly formed bonewithin the defect using the computerized image analysissoftware Image-Pro Plus (IPP) 6.0. Briefly, the image wascorrected for the optical density before the region-of-interest (ROI) was selected in accordance with the colorspecific for the region of new bone formation in thehistological slide. Then the parameters of the ROI, suchas area, were calculated. Automation of these steps isincluded in an algorithm referred as a macro. The macrowas used to normalize the selection of other ROIs. Thepercentage of newly formed bone was determined by ROIarea/total area of images. 5 sections from each sample wereused for semi-quantitative image analysis. 5 regions in eachsection were photographed at 200x magnification. A totalof 25 images for each sample were digitalized.

Immunohistochemical analysis was performed usingthe Histostain-Plus IHC Kit (MR Biotech, Shanghai,CHINA) according to the manufacturer’s instructions.Briefly, tissue samples from defect areas were harvestedfrom 4 groups (four samples per group) at 20 weeks aftertransplantation. The thickness of bone-like tissue in theharvested samples was determined by the Micrometer(126-125 TMC-25, Mitutoyo, Japan). Sections of samples

were prepared as described above, and then deparaffinized,rehydrated and incubated in 3% hydrogen peroxide inmethanol for 10 min to inactivate the endogenousperoxidase. After three times washing with PBS, thesections were digested with Trypsin Diluent (Zhongshan,Beijing, China) for 15 min to unmask antigen binding sitesand then washed three times with PBS. After that, thesections were incubated with 10% goat serum for 20minto block non-specific binding, followed by incubation withthe primary antibody, i.e., rabbit polyclonal antibodyagainst human osteocalcin (OCN, Abcam, Hangzhou,China) overnight at 4°C. Afterwards, the sections werewashed three times with PBS and incubated withbiotinylated IgG against rabbit for 15 min, and thenexposed to horseradish peroxidase (HRP) labeledstreptavidin for 15 min. After washing three times withPBS, the antibody complex was visualized by addition ofa buffered diaminobenzidine (DAB) solution. Then, thesections were counterstained in hematoxylin for 5 min andmounted with neutral gum. The sections treated only withthe second antibody (without primary antibody treatment)were used as a negative control. Finally, the sections werephotographed. To evaluate the protein content of humanOCN expressed by implanted cells in the graft area, weused IPP as the method of semi-quantitative image analysison immunohistochemical section image to measure theIntegrated Optical Density (IOD) (Cybernetics, 2002). Asabove, a total of 25 images per sample were used forcalculation.

StatisticsExperimental results were expressed as means ± standarddeviation (StDev) of the means. All collected data wereexamined by multifactorial analysis of variance.Differences between the independent variables werechecked in post hoc tests (Tukey’s studentized range (HSD)tests for variables). All tests were two-tailed and statisticalsignificance was accepted at P<0.05.

Results

Growth of hMSCs in PLGA scaffoldsThe proliferation of cells in the PLGA scaffold wasevaluated by determining the double-stranded DNAcontent. After the static culture for 24h, the cell density ineach PLGA scaffold was detected as (0.42 ± 0.08) ×106cells/scaffold. During perfusion culture, the cell densityof hMSCs in PLGA scaffold increased gradually andreached a cell density of (2.27 ± 0.27) × 106 cells/scaffoldafter culturing for 12 days, 5-fold the initial cell density.Then, the increase in rate decreased after 12 days. Thereis no significant difference in cell density between day 12and day 15 (P > 0.05) (Fig. 1A). Therefore, perfusionculture of hMSCs in the scaffold for 12 days in vitro wasused to prepare hMSC Construct so as to assure a sufficientcell density and viability for transplantation.

The cell density in PLGA scaffolds after 9 days ofperfusion culture with growth medium was (1.91 ± 0.17)× 106 cell/scaffold. After 14 days of culture with osteogenic



medium for preparation of Osteoblast Construct, the finalcell density reached (2.33 ± 0.31) × 106 cells/scaffold,which indicated a small increase of cell number afterculture with the osteogenic medium and no significantdifference in cell density between Osteoblast Constructand hMSC Construct (P > 0.05) (Fig. 1B).

Osteogenic differentiation of hMSCs in PLGAscaffoldsThe alkaline phosphatase (ALP) activity assay was usedto determine the activity of protein specific forosteogenesis. The PLGA scaffolds with the same densityof standard osteoblasts (HCO4600, Yuhengfeng, Beijing,China) and hMSCs were used as positive and negativecontrols, respectively. The results showed that the level ofALP activity in the Osteoblast Construct (64.5 ± 7.4 μmol/scaffold) was higher than that in the PLGA scaffold with

hMSCs (10.7 ± 2.7 μmol/scaffold) (P < 0.01) (Fig. 2A).There was no significant difference in ALP activitybetween the Osteoblast Construct and the PLGA scaffoldwith standard osteoblasts (72.3 ± 9.9 μmol/scaffold) (P >0.05). RT-PCR was used to detect the gene expression ofosteogenic markers. As shown in Fig. 2B, ALP, COL Iand OPN were expressed in the Osteoblast Constructsimilar to the PLGA scaffold with the standard osteoblasts,while the PLGA scaffold with undifferentiated hMSCsshowed a negative result. Therefore, the above resultsshowed that hMSCs in PLGA scaffolds could be inducedinto osteoblast-like cells after perfusion culture for 14 dayswith the osteogenic medium.

Macroscopic view post-surgeryThe animals were sacrificed respectively at 10 and 20weeks after transplantation and the whole calvaria were

Fig. 1. Proliferation of cells in PLGA scaffold in the perfusion culture system. The cell density in scaffold and the cellproliferation were evaluated by determining the double-stranded DNA (dsDNA) content. (A) Cell densities of hMSCsin PLGA scaffold for culture of 0, 6, 9, 12 and 15 days. (B) Cell densities of hMSCs in PLGA scaffold for culture of0 and 9 days, and then for induction of 14 days with the osteogenic medium. All the experiments were conductedindependently (n = 3). Error bars represent SD. The asterisks (**) indicate statistically significant differences betweenthe relative culture times (P < 0.01).

A B

A B

Fig. 2. Osteogenesis of hMSCs in Osteoblast Construct in vitro. (A) Total alkaline phosphatase (ALP) activity perscaffold. The ALP activity level was compared between scaffolds with hMSCs, osteoblast-like cells differentiatedfrom hMSCs and standard osteoblasts. All the experiments were conducted independently (n = 3). Error bars representSD. The asterisks (**) indicate statistically significant differences between different constructs (P < 0.01). (B)Osteoblast marker gene expression of cells in PLGA scaffold. Lane1: Osteoblast Construct. Lane 2: Scaffold withstandard osteoblasts as a positive control. Lane 3: hMSC Construct as a negative control.



Fig. 3. Macroscopic view of rat calvarias after transplantation. (A)-(D) The graft areas of calvaria at 10 weeks aftertransplantation. (E)-(H) The graft area of calvaria at 20 weeks after transplantation. The left defect of calvaria in (A)and (E) was grafted with Blank Defect while the right defect was grafted with the hMSC Construct. The left defectof calvaria in (B) and (F) was grafted with Blank Scaffold while the right defect was grafted with the hMSC Construct.The left defect of calvaria in (C) and (G) was grafted with Blank Scaffold and the right defect was grafted with theOsteoblast Construct. The left defect of calvaria in (D) and (H) was grafted with the hMSC Construct and the rightdefect was grafted with the Osteoblast Construct. Scale bar: 5mm.

Fig. 4. Cell tracing in vivo at 10 weeks after transplantation. Cells in PLGA scaffold were labeled by fluorescentcarbocyanine CM-Dil. (A) and (B) are staining results of hMSCs and osteoblast-like cells differentiated from hMSCsrespectively. The red staining represents human cells survival in rats after transplantation. (C) and (D) arecounterstaining results with DAPI. The blue staining show native cells in rats.



Blank defect Blank scaffold hMSC Construct Osteoblast Construct

Fig. 5. Histological and immunohistochemical analysis of graft areasafter transplantation. Defect areas were treated with Blank Defect (A),Blank Scaffold (B), hMSCs Construct (C) and Osteoblast Construct(D). 1 series: General images (A1-D1) at 10 weeks after transplantation,2 series: Masson’s Trichrome Staining images (A2-D2) showing thevascularization in the graft area at 10 weeks after transplantation. 3series: General images (A3-D3) showing the bone formation at 20 weeksafter transplantation. 4 series: Images (A4-D4) showing collagenproduced in the newly formed tissues by Masson’s Trichrome Stainingat 20 weeks after transplantation. 5 series: Images (A5-D5) showinghuman osteocalcin (OCN) in the graft areas by immunohistochemistryassay at 20 weeks after transplantation. HB: host bone. NB: new bone.The thin black arrow: the formatted membrane. The thick black arrow: the residual scaffold material. The thin whitearrow: the positive staining with human OCN. The thick white arrow: the inflammatory cells. The white oval: the areaof vascularization. Scale bars A1-D1, A3-D3: 500μm; A2-D2, A4-D4, A5-D5: 100μm. (E) IOD of immunohistochemicalimage against human OCN in the area grafted with hMSC Construct and Osteoblast Construct. IOD of the imagesrepresent the protein content in the area. The error bar represents SD. The asterisks (**) indicate a statistically significantdifference of IOD between the graft area with hMSC Construct and the graft area with Osteoblast Construct (P < 0.01,n=8).

E



harvested for macroscopic examination. Both at 10 weeksand at 20 weeks after transplantation, we found only athin membrane in the defect area without any graft (BlankDefect; Fig. 3A and E). However, fibrous-like connectivetissue was observed in the defect area grafted with BlankScaffold, hMSC Construct or Osteoblast Construct at 10weeks after transplantation. In the graft areas, most of thescaffold materials were degraded and absorbed. Only afew small scaffold grains remained in the graft area (Fig.3B-D). In a comparison of three grafts, the fibrous-likeconnective tissue filled the whole area grafted with hMSCConstruct or Osteoblast Construct. However, the fibrous-like connective tissue was observed only at the edge ofthe area grafted with Blank Scaffold. A thin membraneremained in the central region of the graft area. Nosignificant difference in the morphology of the fibrous-like tissue was observed between the areas grafted withhMSC Construct and Osteoblast Construct. At 20 weeksafter transplantation, scaffold materials were no longerobserved in any of the graft areas (Fig. 3F-H). Bone-liketissue filled in the areas grafted with both the hMSCConstruct and the Osteoblast Construct. However, thebone-like tissue in the area grafted with hMSC Construct(1.23 ± 0.12 mm) was thicker than the tissue in the areagrafted with Osteoblast Construct (0.99 ± 0.14 mm) (P <0.05). In comparison with the areas grafted with hMSCConstruct and Osteoblast Construct, the area grafted withBlank Scaffold was still filled with a fibrous-like tissue.These results showed that both hMSC Construct andOsteoblast Construct had the ability to enhance thereconstruction of calvarial defects of rats.

Cell tracing in vivoAt 10 weeks after transplantation, the graft area washarvested for cell tracing analysis. Under fluoresceinmicroscopy, we detected numerous labeled hMSCs andhMSCs-derived osteoblast-like cells in the graft area. Byvisual inspection, these cells seemed to have aninhomogeneous distribution in the graft area, and thedistribution of osteoblast-like cells seemed to be a littlemore abundant than that of hMSCs (Fig. 4A and B).Counterstaining with DAPI showed that the space withouthuman cells was filled with native cells of rat (blue staining)(Fig. 4C and D), and the distribution of native cells in thearea grafted with hMSC Construct seemed to be moreabundant than that in the area grafted with OsteoblastConstruct.

Histological analysisThe reconstruction process of bone in the calvarial defectwas evaluated by histological analysis. Representativehistological images are showed in Fig. 5. After 10 weeksof transplantation, the defect area without any graft (BlankDefect) was quite different from other graft areas, beinglack of any bone ingrowth. Fig. 5A1 showed only a thinmembrane in the defect area without any graft. However,the connective tissue was produced in the area grafted withBlank Scaffold, hMSC Construct or Osteoblast Construct.The scaffolds had been greatly degraded except for someresidual scaffold materials scattered in the graft area.

However, there was still a difference in scaffolddegradation between areas grafted with these three grafts.Scaffold materials remained more in the area grafted withBlank Scaffold than in the area grafted with hMSCConstruct or Osteoblast Construct. In addition, Fig. 5B1-D1 reflect the difference in inflammatory reaction in thegraft areas. In comparison to areas grafted with BlankScaffold, hMSC Construct and Osteoblast Construct, nouninucleate macrophage and multinucleated inflammatorycells were found in the area grafted with Blank Scaffoldand hMSC Construct, but the area grafted with OsteoblastConstruct showed some inflammatory cells around thescaffold material. In addition, some vessel-like tissue wasdetected in the areas grafted with hMSC Construct,Osteoblast Construct and Blank Scaffold by Masson’sTrichome staining (Fig. 5B2-D2), and the area grafted withhMSC Construct appeared to show a superior vessel-likestructuralization.

At 20 weeks after transplantation, there was no obviouschange in the defect area without any graft (Fig. 5A3) incomparison with that at 10 weeks after transplantation.No classical bone formation was observed in the areagrafted with Blank Scaffold, and some scaffold materialsstill remained in the graft area (Fig. 5B3). However, noresidual scaffold material was detected in both areas graftedwith hMSC Construct and Osteoblast Construct (Fig. 5C3-D3). The obvious small lump bone was formed in thesegraft areas, especially in the area grafted with hMSCConstruct. In comparison with two areas grafted withhMSC Construct and Osteoblast Construct, the area graftedwith hMSC Construct showed some larger woven bone.In addition, collagen, a major extracellular matrix, wasproduced in the newly formed tissues, as determined byMasson’s Trichrome Staining (Fig. 5B4-D4). The bonereconstruction percentage analysis was performed by

Fig. 6. Bone regeneration percentage of the defect areaat 20 weeks after transplantation. The three barsrepresent the different bone regeneration ability of BlankScaffold, hMSC Construct and Osteoblast Construct. Astatistically significant difference was shown (P < 0.01,n=8) between hMSC Construct and OsteoblastConstruct or between hMSC Construct and BlankScaffold (P < 0.01, n=8).



Image-Pro Plus. Fig. 6 indicated that the new boneformation rate in the areas grafted with hMSC Construct(53.9 ± 6.2%) and Osteoblast Construct (41.8 ± 2.5%) washigher than that in the area grafted with Blank Scaffold(20.1 ± 2.1%) (P < 0.01). In addition, there was also asignificant difference in new bone formation percentagebetween the area grafted with hMSCs Construct and thearea grafted with Osteoblast Construct (P < 0.01).

The immunohistochemical analysis specific to humancells was performed to evaluate the survival and functionof hMSC and hMSC-derived osteoblast-like cells in thexenogenic transplantation. The area without any graft orwith Blank Scaffold did not show the positive staining ofhuman osteocalcin (Fig. 5A5 and 5B5). The areas graftedwith hMSC Construct and Osteoblast Construct showedthe positive staining of human osteocalcin (Fig. 5C5 and5D5). However, in comparison of staining images forhuman osteocalcin between Fig. 5C5 and Fig. 5D5, theexpression level of human osteocalcin in the area graftedwith Osteoblast Construct was higher than that in the areagrafted with hMSC Construct (Fig. 5E, P < 0.01). Thenegative control treated only with the second antibodyshowed no any positive staining (no data shown).

Discussion

Autologous transplantation is an optimal choice of bonerepairing treatment. However, xenogenic transplantationmay also have a practical significance in future clinicalapplications, especially in the bone repairing treatment foraged patients. Aged patients with tissue defects may needan allograft transplantation surgery typically because theproliferation and differentiation capacities of stem cellsdecline in an age-dependent way and may even nearlydisappear (Hermann et al., 2010). Therefore, xenogenictransplantation may be an optimal choice to perform tissuereconstruction for those aged recipients in the future.Niemeyer et al. (Niemeyer et al., 2010b) reported thetransplantation of human mesenchymal stem cells in a non-autogenous setting for bone regeneration in a rabbit critical-size defect model. In the present study, the xenogenictransplantation experiment was performed to determinethe effect of hMSCs and osteoblast-like cells differentiatedfrom hMSCs on the reconstruction of calvarial defects inrats, especially to compare the bone reconstruction abilityof undifferentiated hMSCs with that of osteoblast-like cellsdifferentiated from hMSCs in the xenogenictransplantation model. As a prerequisite for comparisonof bone reconstruction abilities between hMSC Constructand Osteoblast Construct, the initial cell densities in thesetwo constructs should be equal. During induction ofhMSCs in PLGA scaffolds into osteoblast-like cells, wefound that cells still increased to a small extent. Therefore,to decrease the difference in cell densities between hMSCConstruct and Osteoblast Construct as much as possible,hMSC Construct was prepared by undergoing 12-day-culture in the growth medium, whereas OsteoblastConstruct was prepared as that the PLGA scaffold withhMSCs undergoing 9-day-culture in the growth medium

was continued in the osteogenic medium for another 14days. The cell number increased during 14-day-inductionwhich could compensate for the difference in cell densitiesbetween scaffolds for 9-day-cultures and 12-day-culturesin the growth medium.

Critical size defect (CSD) experimental models areessential for in vivo experiments for bone reconstruction.The calvarial CSD of the rat is one of the models for newbone reconstruction and especially a good model for flatbone tissue engineering because the calvaria are suitablefor the creation of defects, implantation of grafts, andanalysis of reconstruction (Hollinger and Kleinschmidt,1990; Sweeney et al., 1995). In previous studies, the smalldefect in 3mm diameter could undergo spontaneous boneregeneration. However, the defect in 5mm diameter isbeyond the size of spontaneous bone regeneration (AybarOdstrcil et al., 2005; Wurzler et al., 1998). Therefore, the5mm diameter of bilateral full-thickness calvarial defectsin 2-month-old SD rats, a proper size of CSD proved byBosch (Bosch et al., 1998), were used in the presenttransplantation experiment. In the present experiment, nosignificant bone formation except for a thin membrane wasobserved in the defect area without any graft. Thisdemonstrated that the 5mm diameter defect is beyond thesize of spontaneous bone reconstruction.

Some studies have shown that PLGA could supportMSC proliferation and differentiation both in vitro and invivo (Ishaug-Riley et al., 1997; Karp et al., 2003), and theporous PLGA scaffolds have been utilized for thereconstruction of bone including mandible and diaphysis(Ren et al., 2005; Yoon et al., 2007). We also used thePLGA scaffold without any cells as a control to assess thepromotion effect of hMSCs or osteoblast-like cellsdifferentiated from hMSCs on the bone reconstruction ofrat calvaria. The PLGA scaffold used in the presentexperiment was fabricated without any xenogenic growthfactors so as to reflect really the bone reconstruction abilityof hMSCs or osteoblast-like cell differentiated fromhMSCs. As in the previous report (Ren et al., 2005), thePLGA scaffold alone could not reconstruct the whole largerdefect area except for formation of bone-like tissue at therim of defect. The porous scaffold is only a carrier for cellcreeping and has no bone inducement function. The keyaim of our experiment was to compare the bonereconstruction capacities between undifferentiated hMSCsand osteoblast-like cells differentiated from hMSCs.Significant bone formation was observed in the area graftedwith hMSC Construct or Osteoblast Construct. Thethickness of new bone formed in the graft area was thinnerthan that of rat calvaria surrounding the graft area, but thenew bone in the graft area was relatively compact. As inthe macroscopic view and histological assay describedabove, the undifferentiated hMSCs seemed to have asuperior ability of bone reconstruction in comparison withthat of osteoblast-like cells differentiated from hMSCs.According to the histological assay, the new boneformation in the area grafted with hMSC Construct wassuperior to that in the area grafted with OsteoblastConstruct. This was confirmed by the statistical analysisof bone formation. Of course, micro CT is a standard tool



to quantify bone regeneration. Regretfully, there was nothe micro CT instrument available to measurequantitatively the new bone formation. Some othermethods such as macroscopic examination, histologicalassay and semi-quantitative bone formation analysis,however, may be acceptable for evaluation of bonereconstruction in rat calvaria.

Our result that undifferentiated hMSCs induced morenew bone formation in the graft area seems in contradictionto a previous study on the athymic mice transplantationmodel (Meinel et al., 2005). According to the analysis ofcell tracing in vivo, both hMSC and osteoblast-like cellsdifferentiated from hMSCs survived in rats. However, theabundance of hMSCs seemed to be less than that ofosteoblast-like cells. It should be inferred that the effectof the hMSCs on the promotion of bone reconstruction ofxenogenic animals lies not only on the osteogenicdifferentiation of hMSCs to reconstruct the defect directly(Kim et al., 2007), but also on their indirect function inthe xenogenic transplantation, e.g., hMSCs may secretsome growth factors to recruit native cells into the defectarea to generate new osseous tissues (Bi et al., 1999; Franket al., 2002). As shown in Fig. 4, there were native cells(in blue) surrounding the transplanted cells (in red). It issupposed that these native cells are translocated from thenative environment surrounding the graft area. In addition,the distribution of native cells in the defect area graftedwith hMSC Construct appeared to be more abundant incomparison with that in the defect area grafted withOsteoblast Construct. If this difference had been verifiedby further studies, we could infer that the hMSCs shouldhave superior ability of recruiting native cells comparedto osteoblast-like cells. Of course, it is also possible thatthe signal shown in Fig. 4 is from macrophages digestinglabeled hMSCs or rat cells labeled due to in vivo labeltransfer. This possibility should be explored in furtherstudies.

The above inference should also be confirmed by ourimmunohistochemistry assay of human osteocalcin.Although the positive staining rate of osteocalcin fromhuman in the area grafted with Osteoblast Construct washigher than that in the area grafted with hMSC Construct,the total bone reconstruction in the area grafted with hMSCConstruct was superior to that in the area grafted withOsteoblast Construct. In addition, stem cells may feelsignals from the microenvironment and thus differentiateinto vascular endothelial cells, or send out signals to recruitendothelial progenitor cells from the circulating blood.These effects are contributed to angiogenesis that is oneof the most important factors for bone formation (Kim etal., 2007). However, osteoblast-like cells differentiatedfrom hMSCs may lose the ability to differentiate intoendothelial cells and hence lose the potential forangiogenesis. As shown in Fig. 5C2 and D2, a vessel-likestructure in the area grafted with hMSCs Construct andOsteoblast Construct was detected, and the formation ofthe vessel-like structure in the area grafted with hMSCConstruct was superior to that in the area grafted withOsteoblast Construct. However, it should be verified bythe vascular specific staining in a further study whether

the formation of a vessel-like structure in the graft area isindeed a sign of angiogenesis.

The immunogenicity of allogeneic cells in the recipientis an important problem in clinical cell transplantation.Previous studies demonstrated that both undifferentiatedand differentiated MSCs have low immunogenic profiles(Le Blanc et al., 2003a; Le Blanc et al., 2003b;Pierdomenico et al., 2005; Niemeyer et al., 2007). Thosestudies showed that MSCs do not elicit alloreactivelymphocyte responses in vitro due to their special immunecharacteristics. Nevertheless, Poncelet et al. (Poncelet etal., 2007) found that xenogenic MSCs with lowimmunogenicity in vitro elicited an immune response invivo after transplantation. In our experiments, aninflammatory reaction, such as production of macrophages,was not observed in the area grafted with hMSC Construct.However, this reaction was elicited in the area grafted withOsteoblast Construct, especially in the initial period aftertransplantation. Though this is not enough to prove thepresentation of a local immune response, it can be anindication that a local inflammation occurred. The effectof the PLGA scaffold on the inflammatory reaction wasexcluded since the same reaction was not elicited in thearea grafted with Blank Scaffold. Therefore, it is deducedthat the immunogenic characteristic of osteoblast-like cellsdifferentiated from hMSCs should be different from thatof hMSCs or that the immunogenic characteristics ofosteoblast-like cells should have changed in vivo aftertransplantation. The difference of immunogeniccharacteristics in vitro and in vivo between hMSCs andosteoblast-like cells differentiated from hMSCs should bea subject for further research.

Conclusion

In conclusion, the present study demonstrates that hMSCenable bone reconstruction of calvarial defects in axenogenic transplantation model and may have superiorpotential of bone reconstruction compared to osteoblast-like cells according to morphological observations andhistochemical analysis.

Acknowledgements

We would like to thank three anonymous reviewers forcritical comments on this manuscript. This study wasfinancially supported by Scientific Research fromScientific Fund of Zhejiang (2009C13020) and NationalNatural Science Fund of China (30971460).

References

Ahmad S, Ahmed A (2004) Elevated placental solublevascular endothelial growth factor receptor-1 inhibitsangiogenesis in preeclampsia. Circ Res 95: 884-891.

Aybar Odstrcil A, Territoriale E, Missana L (2005) Anexperimental model in calvaria to evaluate bone therapies.Acta Odontol Latinoam 18: 63-67.

Bernardo ME, Avanzini MA, Perotti C, Cometa AM,Moretta A, Lenta E, Del Fante C, Novara F, de Silvestri A,



Amendola G, Zuffardi O, Maccario R, Locatelli F (2007)Optimization of in vitro expansion of human multipotentmesenchymal stromal cells for cell-therapy approaches:further insights in the search for a fetal calf serumsubstitute. J Cell Physiol 211: 121-130.

Bi LX, Simmons DJ, Mainous E (1999) Expression ofBMP-2 by rat bone marrow stromal cells in culture. CalcifTissue Int 64: 63-68.

Bosch C, Melsen B, Vargervik K (1998) Importanceof the critical-size bone defect in testing bone-regeneratingmaterials. J Craniofac Surg 9: 310-316.

Bruder SP, Kraus KH, Goldberg VM, Kadiyala S(1998a) The effect of implants loaded with autologousmesenchymal stem cells on the healing of canine segmentalbone defects. J Bone Joint Surg Am 80: 985-996.

Bruder SP, Kurth AA, Shea M, Hayes WC, Jaiswal N,Kadiyala S (1998b) Bone regeneration by implantation ofpurified, culture-expanded human mesenchymal stemcells. J Orthop Res 16: 155-162.

Cowan CM, Shi YY, Aalami OO, Chou YF, Mari C,Thomas R, Quarto N, Contag CH, Wu B, Longaker MT(2004) Adipose-derived adult stromal cells heal critical-size mouse calvarial defects. Nat Biotechnol 22: 560-567.

Cui L, Liu B, Liu G, Zhang W, Cen L, Sun J, Yin S,Liu W, Cao Y (2007) Repair of cranial bone defects withadipose derived stem cells and coral scaffold in a caninemodel. Biomaterials 28: 5477-5486.

Cybernetics M (2002) Image-Pro Plus – applicationnotes. Media Cybernetics, Silver Spring, MD.

Ferrari A, Hannouche D, Oudina K, Bourguignon M,Meunier A, Sedel L, Petite H (2001) In vivo tracking ofbone marrow fibroblasts with fluorescent carbocyaninedye. J Biomed Mater Res 56: 361-367.

Frank O, Heim M, Jakob M, Barbero A, Schafer D,Bendik I, Dick W, Heberer M, Martin I (2002) Real-timequantitative RT-PCR analysis of human bone marrowstromal cells during osteogenic differentiation in vitro. JCell Biochem 85: 737-746.

Garvey W (1984) Modified elastic tissue – Massontrichrome stain. Stain Technol 59: 213-216.

Gong Z, Niklason LE (2008) Small-diameter humanvessel wall engineered from bone marrow-derivedmesenchymal stem cells (hMSCs). FASEB J 22: 1635-1648.

Gopferich A (1996) Mechanisms of polymerdegradation and erosion. Biomaterials 17: 103-114.

Hao W, Pang L, Jiang M, Lv R, Xiong Z, Hu YY (2010)Skeletal repair in rabbits using a novel biomimeticcomposite based on adipose-derived stem cellsencapsulated in collagen I gel with PLGA-beta-TCPscaffold. J Orthop Res 28: 252-257.

Hermann A, List C, Habisch HJ, Vukicevic V, Ehrhart-Bornstein M, Brenner R, Bernstein P, Fickert S, Storch A(2010) Age-dependent neuroectodermal differentiationcapacity of human mesenchymal stromal cells: limitationsfor autologous cell replacement strategies. Cytotherapy 12:17-30.

Hollinger JO, Kleinschmidt JC (1990) The critical sizedefect as an experimental model to test bone repairmaterials. J Craniofac Surg 1: 60-68.

Holtorf HL, Jansen JA, Mikos AG (2006) Modulationof cell differentiation in bone tissue engineering constructscultured in a bioreactor. Adv Exp Med Biol 585: 225-241.

Ishaug-Riley SL, Crane GM, Gurlek A, Miller MJ,Yasko AW, Yaszemski MJ, Mikos AG (1997) Ectopic boneformation by marrow stromal osteoblast transplantationusing poly(DL-lactic-co-glycolic acid) foams implantedinto the rat mesentery. J Biomed Mater Res 36: 1-8.

Jorgensen C, Gordeladze J, Noel D (2004) Tissueengineering through autologous mesenchymal stem cells.Curr Opin Biotechnol 15: 406-410.

Karp JM, Shoichet MS, Davies JE (2003) Boneformation on two-dimensional poly(DL-lactide-co-glycolide) (PLGA) films and three-dimensional PLGAtissue engineering scaffolds in vitro. J Biomed Mater ResA 64: 388-396.

Kassem M, Risteli L, Mosekilde L, Melsen F, EriksenEF (1991) Formation of osteoblast-like cells from humanmononuclear bone marrow cultures. APMIS 99: 269-274.

Kim J, Kim IS, Cho TH, Lee KB, Hwang SJ, Tae G,Noh I, Lee SH, Park Y, Sun K (2007) Bone regenerationusing hyaluronic acid-based hydrogel with bonemorphogenic protein-2 and human mesenchymal stemcells. Biomaterials 28: 1830-1837.

Koc ON, Gerson SL, Cooper BW, Dyhouse SM,Haynesworth SE, Caplan AI, Lazarus HM (2000) Rapidhematopoietic recovery after coinfusion of autologous-blood stem cells and culture-expanded marrowmesenchymal stem cells in advanced breast cancer patientsreceiving high-dose chemotherapy. J Clin Oncol 18: 307-316.

Kon E, Muraglia A, Corsi A, Bianco P, Marcacci M,Martin I, Boyde A, Ruspantini I, Chistolini P, Rocca M,Giardino R, Cancedda R, Quarto R (2000) Autologousbone marrow stromal cells loaded onto poroushydroxyapatite ceramic accelerate bone repair in critical-size defects of sheep long bones. J Biomed Mater Res 49:328-337.

Korda M, Blunn G, Goodship A, Hua J (2008) Use ofmesenchymal stem cells to enhance bone formation aroundrevision hip replacements. J Orthop Res 26: 880-885.

Kubler N, Michel C, Zoller J, Bill J, Muhling J, ReutherJ (1995) Repair of human skull defects usingosteoinductive bone alloimplants. J Craniomaxillofac Surg23: 337-346.

Langer R, Vacanti JP (1993) Tissue engineering.Science 260: 920-926.

Lazarus HM, Koc ON, Devine SM, Curtin P, MaziarzRT, Holland HK, Shpall EJ, McCarthy P, Atkinson K,Cooper BW, Gerson SL, Laughlin MJ, Loberiza FR Jr,Moseley AB, Bacigalupo A (2005) Cotransplantation ofHLA-identical sibling culture-expanded mesenchymalstem cells and hematopoietic stem cells in hematologicmalignancy patients. Biol Blood Marrow Transplant 11:389-398.

Le Blanc K, Tammik C, Rosendahl K, Zetterberg E,Ringden O (2003a) HLA expression and immunologicproperties of differentiated and undifferentiatedmesenchymal stem cells. Exp Hematol 31: 890-896.



Le Blanc K, Tammik L, Sundberg B, Haynesworth SE,Ringden O (2003b) Mesenchymal stem cells inhibit andstimulate mixed lymphocyte cultures and mitogenicresponses independently of the major histocompatibilitycomplex. Scand J Immunol 57: 11-20.

Mankani MH, Kuznetsov SA, Wolfe RM, MarshallGW, Robey PG (2006) In vivo bone formation by humanbone marrow stromal cells: reconstruction of the mousecalvarium and mandible. Stem Cells 24: 2140-2149.

Meinel L, Fajardo R, Hofmann S, Langer R, Chen J,Snyder B, Li C, Zichner L, Langer R, Vunjak-NovakovicG, Kaplan DL (2005) Silk implants for the healing ofcritical size bone defects. Bone 37: 688-698.

Miura M, Miura Y, Sonoyama W, Yamaza T, GronthosS, Shi S (2006) Bone marrow-derived mesenchymal stemcells for regenerative medicine in craniofacial region. OralDis 12: 514-522.

Niemeyer P, Kornacker M, Mehlhorn A, Seckinger A,Vohrer J, Schmal H, Kasten P, Eckstein V, Südkamp NP,Krause U (2007) Comparison of immunological propertiesof bone marrow stromal cells and adipose tissue-derivedstem cells before and after osteogenic differentiation invitro. Tissue Eng 13: 111-121.

Niemeyer P, Schonberger TS, Hahn J, Kasten P,Fellenberg J, Suedkamp N, Mehlhorn AT, Milz S, PearceS (2010a) Xenogenic transplantation of humanmesenchymal stem cells in a critical size defect of the sheeptibia for bone regeneration. Tissue Eng Part A 16: 33-43.

Niemeyer P, Szalay K, Luginbuhl R, Südkamp NP,Kasten P (2010b) Transplantation of human mesenchymalstem cells in a non-autogenous setting for boneregeneration in a rabbit critical-size defect model. ActaBiomater 6: 900-908.

Pierdomenico L, Bonsi L, Calvitti M, Rondelli D,Arpinati M, Chirumbolo G, Becchetti E, Marchionni C,Alviano F, Fossati V, Staffolani N, Franchina M, GrossiA, Bagnara GP (2005) Multipotent mesenchymal stem cellswith immunosuppressive activity can be easily isolatedfrom dental pulp. Transplantation 80: 836-842.

Poncelet AJ, Vercruysse J, Saliez A, Gianello P (2007)Although pig allogeneic mesenchymal stem cells are notimmunogenic in vitro, intracardiac injection elicits animmune response in vivo. Transplantation 83: 783-790.

Ren T, Ren J, Jia X, Pan K (2005) The bone formationin vitro and mandibular defect repair using PLGA porousscaffolds. J Biomed Mater Res A 74: 562-569.

Sweeney TM, Opperman LA, Persing JA, Ogle RC(1995) Repair of critical size rat calvarial defects usingextracellular matrix protein gels. J Neurosurg 83: 710-715.

Wurzler KK, DeWeese TL, Sebald W, Reddi AH (1998)Radiation-induced impairment of bone healing can beovercome by recombinant human bone morphogeneticprotein-2. J Craniofac Surg 9: 131-137.

Xie CG, Wang JF, Xiang Y, Jia BB, Qiu LY, Wang LJ,Wang GZ, Huang GP (2005) Marrow mesenchymal stemcells transduced with TPO/FL genes as support for ex vivoexpansion of hematopoietic stem/progenitor cells. Cell MolLife Sci 62: 2495-2507.

Yamada Y, Ueda M, Naiki T, Takahashi M, Hata K,Nagasaka T (2004) Autogenous injectable bone forregeneration with mesenchymal stem cells and platelet-rich plasma: tissue-engineered bone regeneration. TissueEng 10: 955-964.

Yang J, Cao C, Wang W, Tong X, Shi D, Wu F, ZhengQ, Guo C, Pan Z, Gao C, Wang J (2010) Proliferation andosteogenesis of immortalized bone marrow-derivedmesenchymal stem cells in porous polylactic glycolic acidscaffolds under perfusion culture. J Biomed Mater Res A93: 817-829.

Yoon SJ, Park KS, Kim MS, Rhee JM, Khang G, LeeHB (2007). Repair of diaphyseal bone defects withcalcitriol-loaded PLGA scaffolds and marrow stromal cells.Tissue Eng 13: 1125-1133.

Younger EM, Chapman MW (1989) Morbidity at bonegraft donor sites. J Orthop Trauma 3: 192-195.

Zhou QL, Gong YH, Gao CY (2005) Microstructureand mechanical properties of poly(L-lactide) scaffoldsfabricated by gelatin particle leaching method. J ApplPolymer Sci 98: 1373-1379.

Zsarnovszky A, Le HH, Wang HS, Belcher SM (2005)Ontogeny of rapid estrogen-mediated extracellular signal-regulated kinase signaling in the rat cerebellar cortex:potent nongenomic agonist and endocrine disruptingactivity of the xenoestrogen bisphenol A. Endocrinology146: 5388-5396.

RECONSTRUCTION OF RAT CALVARIAL DEFECTS WITH HUMAN ... · C Zong European Cells and Materials Vol....

Documents

Transcript of RECONSTRUCTION OF RAT CALVARIAL DEFECTS WITH HUMAN ... · C Zong European Cells and Materials Vol....